Content-Based Conflict of Interest Detection on Wikipedia

Content-Based Conflict of Interest Detection on Wikipedia Udochukwu Orizu, Yulan He School of Engineering and Applied Science, Aston University, UK {orizuus,y.he9}@aston.ac.uk Abstract Wikipedia is one of the most visited websites in the world. On Wikipedia, Conflict-of-Interest (CoI) editing happens when an editor uses Wikipedia to advance their interests or relationships. This includes paid editing done by organisations for public relations purposes, etc. CoI detection is highly subjective and though closely related to vandalism and bias detection, it is a more difficult problem. In this paper, we frame CoI detection as a binary classification problem and explore various features which can be used to train supervised classifiers for CoI detection on Wikipedia articles. Our experimental results show that the best F-measure achieved is 0.67 by training SVM from a combination of features including stylometric, bias and emotion features. As we are not certain that our non-CoI set does not contain any CoI articles, we have also explored the use of one-class classification for CoI detection. The results show that using stylometric features outperforms other types of features or a combination of them and gives an F-measure of 0.63. Also, while binary classifiers give higher recall values (0.81∼0.94), one-class classifier attains higher precision values (0.69∼0.74). Keywords: Wikipedia, Conflict-of-Interest Detection, Bias, Stylometric features, One-class classification. 1. Introduction A key feature of Wiki sites is to allow people from all over the world to add or modify articles anonymously and with- out consequence. This enables people with malicious inten- tions to use articles to promote or to discredit target prod- ucts, services, organisations, or individuals. Conflict of Interest (CoI) is defined as a situation in which a person or organisation is involved in multiple interests, financial interest, or otherwise; one of which could pos- sibly corrupt the motivation of the individual or organisa- tion1. According to Wikipedia, content on Wikipedia and other Wiki-media projects “must be written from a neutral point of view (NPOV)”2. NPOV refers to representing neu- tral and without bias all of the significant views that have been published by reliable sources. CoI editing happens when an editor contributes to Wikipedia about themselves or their relationships such as family, friends, clients, em- ployers, and financial links, etc. Often times, CoI editing does not comply with NPOV. CoI editing is strongly discouraged by Wikipedia as it un- dermines the public’s confidence in it, and causes public embarrassment to the individuals being promoted. It is easy to assume that CoI is just bias; however while it is not pos- sible for CoI to exist without bias, bias can often exist in the absence of a CoI. One’s beliefs and desires can lead to biased editing, but that does not constitute a CoI. The growth of Wikipedia makes it increasingly difficult for both Wikipedia users and administrators to manually monitor articles. Taking two example documents from our dataset, one is an article classified as CoI while the other is not: • CoI example: Kaizaad Kotwa, born in Mumbai, In- dia, is an award winning …

Content truncated to 500 words